This document analyzes the technical, allocative, and economic efficiencies of farmers using the System of Rice Intensification (SRI) method versus conventional rice farming in Tamil Nadu, India. Key findings include: - SRI farms had higher average technical (92%), allocative (76%), and economic (70%) efficiencies compared to conventional farms (73%, 35%, and 25% respectively). - While SRI reduced some input costs like seeds, fertilizer, and water use, total production costs were only 6.4% lower for SRI farms. Increased yields of 21.7% primarily made SRI more profitable. - More research is needed on further reducing SRI costs and

1. K. Palanisami
Director, IWMI-Tata Water Policy Research Programme
IWMI South Asia Regional Office, Hyderabad
&
C. R. Ranganathan Senthilnathan
Tamil Nadu Agricultural University, Coimbatore

2. Data sources
 Farmer Participatory Action Research Program
(FPARP), Ministry of Water Resources
 Location: Thanjavur district of Tamil Nadu
 Primary data from 2007 Rabi season
 60 farmers randomly selected:
 30 SRI, and
 30 Non-SRI

3. Approach
• Econometric model: Stochastic frontier function
• Technical efficiency is the ability to produce
maximum output with a given quantity of inputs.
It is the ratio of actual output to maximum
possible output.
• Allocative efficiency refers to the ability of
choosing optimal input levels for current output
at given factor prices.
• Economic efficiency is the product of technical
and allocative efficiency.

4. Methodology
Stochastic Frontier
The following equation denotes the production frontier in
the matrix form:
Yi = f (X ; β) exp (v
i
- u
i
)
; i = 1, 2, ……….,n
Where:
Yi = the output of the ith
farm
Xi = inputs for the ith
farm
β = the vector of parameters to be estimated
vi = the symmetric component of the error term
ui = the non-negative random variable which is
under the control of the farm

5. Technical Efficiency
Farm-specific estimates of technical efficiency are defined by:
Where is the cumulative function of the standard normal variable
is an estimated parameter of the
conditional distribution
( ){ }
[ ]
[ ] 





+
Φ−
+Φ−
=−=
2
*
*
**
2
1
exp
/1
/1
/exp i
i
ii
ui
ui
uiu
iii uETE σγε
σγε
σγεσ
ε
Φ
2*
)1( εσγγσ −=iu
iiu ε/

6. Allocative Efficiency
The stochastic cost frontier is given by
Where:
ci = the observed cost of production for the ith
farm,
C = the deterministic kernel (such as Cobb-Douglas form),
wi = a vector of prices of input variables,
β = a vector of unknown parameters to be estimated,
vi = a two-sided error term representing statistical noise, and
ui = a non-negative cost-inefficiency effect.
( ) iiiii uvwyCc ++= β;,ln

7. Sample mean of resources used (per ha)
Variables SRI Conventional
%
Difference
Seed rate (kg) 7.5 81.16 -90.8
Fertilizer (NPK in kg) 339.96 367.85 -7.6
Human labour (man-days) 184.88 166.38 +10.8
Water use* (mm) 845 1,240 -32.8
Yield (tons) 6.61 5.43 +21.7
Results and Discussion
* excluding effective rainfall

8. 0
200
400
600
800
1000
1200
1400
Seed rate(Kgs) Fertilizer (NPK in
Kgs)
Human labour
(man days)
Wateruse (mm) Yield (tonnes)
7.5
339.96
184.88
845
6.61
81.16
367.85
166.38
1240
5.43
SRI Conventional
Sample mean of resources used (per ha basis)

9. Particulars SRI Conventional % difference
Seeds & nursery 592 1,515 - 60.9
Human labour 12,242 9,983 +22.6
Machine power 3,495 4,136 - 15.5
Agro chemicals 927 1,698 - 45.0
Fertilizers 3,060 3,311 - 7.6
Manures 1,325 2,466 - 46.3
Total cost 21,640 23,107 - 6.4
Total income 42,965 35,295 +21.7
Net income 21,325 12,188 +75.0
Economics of rice production ( Rs. per ha )

10. Economics of rice production
0
5000
10000
15000
20000
25000
30000
35000
40000
45000
Seeds&Nursery
Humanlabour
MachinePower
AgroChemicals
Fertilizers
Manures
Totalcost
Totalincome
Netincome
591.98
12241.601
3495.45
926.58
3059.64
1325
21640.25
42965
21324.76
1514.5
9982.8
4135.55
1698.16
3310.65
2465.5
23107.16
35295
12187.84
Rs/ha
SRI

11. Distribution of technical, allocative & economic efficiencies
Efficiency
(%)
SRI Conventional
Technical
efficiency
Allocative
efficiency
Economic
efficiency
Technical
efficiency
Allocative
efficiency
Economic
efficiency
0 to 19 - - - 1 (3) - 7 (23)
20-29 - - - 1 (3) 8 (27) 15 (50)
30-39 - - - 0 (0) 17 (57) 8 (27)
40-49 - - - 2 (7) 4 (13) -
50-59 - - 2 (7) 1 (3) 1 (3) -
60-69 3 (10) 9 (30) 3 (10) - -
70-79 1 (3) 23 (77) 18 (60) 12 (40) - -
80-90 6 (20) 4 (13) 1 (3) 5 (17) - -
90-95 20 (67) - - 2 (7) - -
>95 3 (10) - - 3 (10) 0 -
Total 30 (100) 30 (100) 30 (100) 30 (100) 30 (100) 30 (100)
Mean (%) 92 76 70 73 35 25
Minimum (%) 73 67 56 10 26 5
Maximum (%) 98 85 82 99 51 37
Figures in parenthesis denote the percentage to the total number of farmers

12. 0 0 0 0 0 0
1
23
1 1
0
2
1
3
12
5 5
6
0
5
10
15
20
25
10 25 35 45 55 65 75 85 95
Technical Efficiency (per cent)
Numberoffarmers
SRI Conventional
Distribution of Technical Efficiency

13. 0 0 0 0 0
3
23
4
00
8
17
4
1
0 0 0 00
5
10
15
20
25
10 25 35 45 55 65 75 85 95
Allocative Efficiency
NumberofFarmers
SRI Conventional
Distribution of Allocative Efficiency

14. 0 0 0 0
2
9
18
1
0
7
15
8
0 0 0 0 0 00
2
4
6
8
10
12
14
16
18
20
10 25 35 45 55 65 75 85 95
Economic efficiency
NumberofFarmers
SRI Conventional
Distribution of Economic Efficiency

15. Conclusion
 SRI farms are comparatively more
efficient
 Cost reductions are not significant
 Increased yield primarily makes SRI
attractive
 Sustained yield & prices will decide the
future of SRI
 More studies are needed on:
• Cost reductions aspects
• Sustainability aspects